光热电站需要配备大规模高温储热模块, 金属氧化物可以通过可逆氧化还原反应实现热量的存储与释放。其中锰基氧化物无毒、廉价, 极具潜力, 但可逆性较差。为此, 本研究采用深共溶溶剂离子热合成了锰基氧化物, 探索了合成参数和铁掺杂对其储热性能的影响。离子热合成的MnCO₃前驱体在高温下分解释放CO₂, 使锰基氧化物具有丰富的孔隙结构, 为氧气的传输与扩散提供通道, 有利于氧化还原反应。离子热合成的Mn₂O₃比商业Mn₂O₃反应性能好, 但其氧化反应速度较慢; 合成温度150 ℃、掺杂20% Fe的锰铁氧化物的氧化速率快, 储热密度高达300.66 J/g, 反应可逆性最佳, 可实现长期稳定循环。离子热合成策略可以增加锰氧化物中晶格氧占比, 促进氧空位的迁移, 从而提高可逆性和循环稳定性。

Multiple object tracking (MOT) in unmanned aerial vehicle (UAV) videos has attracted attention. Because of the observation perspectives of UAV, the object scale changes dramatically and is relatively small. Besides, most MOT algorithms in UAV videos cannot achieve real-time due to the tracking-by-detection paradigm. We propose a feature-aligned attention network (FAANet). It mainly consists of a channel and spatial attention module and a feature-aligned aggregation module. We also improve the real-time performance using the joint-detection-embedding paradigm and structural re-parameterization technique. We validate the effectiveness with extensive experiments on UAV detection and tracking benchmark, achieving new state-of-the-art 44.0 MOTA, 64.6 IDF1 with 38.24 frames per second running speed on a single 1080Ti graphics processing unit.

针对现有的基于生成对抗网络的图像融合方法存在忽略红外图像的热目标信息，可见光图像的部分细节信息丢失的情况，提出一种基于双路级联对抗机制的红外与可见光图像融合方法。首先，在生成器模型中采用双路级联的方式分别对红外与可见光图像进行特征提取，为提高融合质量在内容损失函数中引入结构相似度。然后，在判别器模型中采用双判别方式判别生成图像的真假。最后，在公开的数据集上进行实验并与8种经典先进的图像融合方法进行比较。实验结果表明：融合图像在保留红外图像的目标信息同时也保留了可见光图像更多的细节信息，在主观评价和客观评估上均获得较好的性能。

By improving the long-term correlation tracking (LCT) algorithm, an effective object tracking method, improved LCT (ILCT), is proposed to address the issue of occlusion. If the object is judged being occluded by the designed criterion, which is based on the characteristic of response value curve, an added re-detector will perform re-detection, and the tracker is ordered to stop. Besides, a filtering and adoption strategy of re-detection results is given to choose the most reliable one for the re-initialization of the tracker. Extensive experiments are carried out under the conditions of occlusion, and the results demonstrate that ILCT outperforms some state-of-the-art methods in terms of accuracy and robustness.